AI Agents Trigger Privacy Backlash as Apple Tightens Controls
The rapid adoption of autonomous AI agents has triggered a wave of privacy concerns, prompting major tech firms to implement stricter security controls. Apple recently updated its macOS operating system to limit AI access to sensitive data like messages and browsing history, citing risks that grow as agents become more autonomous. These measures follow reports of unauthorized data access by Meta's Muse agent, which has faced user backlash for exposing personal information and making pricing errors. Experts warn that even users who do not use AI agents face privacy risks, as their data may be accessed through the agents of their contacts.
Key points
- Apple introduced new macOS controls to restrict AI agents from accessing mail, messages, and browsing history without explicit user consent, responding to growing privacy risks.
- Meta's Muse AI agent faced criticism after a user reported that it disclosed their home address via Facebook Marketplace due to 'always allow' permissions, and it also made a pricing error by accepting a price $100 below the user's minimum.
- Meta acknowledged the Muse pricing error but maintained that its integrations are 'entirely opt-in,' requiring users to enable specific security features like 'Full Disk Access' and a Messages connector.
- Privacy experts warn that consumers who do not use AI agents are not shielded from risks, as agents can read inboxes and texts containing data from friends and family who have not explicitly consented to such access.
- The AI agent boom has led to increased scrutiny, with users like Robb disconnecting Muse from their apps after privacy incidents, though some plan to re-enable the tool as it is fine-tuned.
Background
This development follows earlier concerns about Meta's Muse AI agent, which launched in September 2026 and quickly became the top app in the U.S. despite facing criticism over privacy and security flaws. Previous reports highlighted that Muse built detailed dossiers on users and their contacts, and internal documents revealed rushed fixes for critical vulnerabilities before launch. Additionally, OpenAI notified the city of Chicago in September 2026 that its AI agents accessed a public-facing municipal database, highlighting global concerns over autonomous AI systems acting without authorization. The surge in AI agent adoption has also impacted financial markets, with Charles Schwab's stock falling 6% in September 2026 as investors feared AI agents would disrupt legacy financial intermediaries.
Why it matters
The rise of AI agents has exposed significant gaps in current privacy frameworks, as these tools can access and process sensitive data without users fully understanding the extent of their permissions. This has led to a push for stricter controls from tech companies like Apple, but experts warn that the ripple effects of AI agents accessing data on behalf of users can compromise the privacy of third parties who have not consented to such access. The incident highlights the need for clearer user consent mechanisms and robust security measures to mitigate the risks associated with increasingly autonomous AI systems.
What to watch
Tech companies are expected to continue refining AI agent permissions and security controls in response to privacy concerns. Users may see more granular consent options and stricter default settings for AI agents accessing sensitive data. Additionally, regulatory bodies may scrutinize AI agent data practices, potentially leading to new guidelines or regulations to protect consumer privacy. The fine-tuning of AI agents like Meta's Muse and the launch of new tools like Grok Bot from X will likely be closely monitored for privacy and security improvements.
Want the full story? Read the original reporting
Read on AP News